Confidence-Weighted Bipartite Ranking

TL;DR

This paper introduces a confidence-weighted online bipartite ranking algorithm that improves scalability and reliability in class-imbalanced, high-dimensional data streams, outperforming existing methods in empirical tests.

Contribution

It proposes a novel linear online confidence-weighted bipartite ranking algorithm and its diagonal variant for high-dimensional data, addressing limitations of prior linear ranking methods.

Findings

Outperforms existing online AUC maximization algorithms.

Effective on benchmark and high-dimensional datasets.

Reliable and scalable in class-imbalanced streaming scenarios.

Abstract

Bipartite ranking is a fundamental machine learning and data mining problem. It commonly concerns the maximization of the AUC metric. Recently, a number of studies have proposed online bipartite ranking algorithms to learn from massive streams of class-imbalanced data. These methods suggest both linear and kernel-based bipartite ranking algorithms based on first and second-order online learning. Unlike kernelized ranker, linear ranker is more scalable learning algorithm. The existing linear online bipartite ranking algorithms lack either handling non-separable data or constructing adaptive large margin. These limitations yield unreliable bipartite ranking performance. In this work, we propose a linear online confidence-weighted bipartite ranking algorithm (CBR) that adopts soft confidence-weighted learning. The proposed algorithm leverages the same properties of soft confidence-weighted learning in a framework for bipartite ranking. We also develop a diagonal variation of the proposed confidence-weighted bipartite ranking algorithm to deal with high-dimensional data by maintaining only the diagonal elements of the covariance matrix. We empirically evaluate the effectiveness of the proposed algorithms on several benchmark and high-dimensional datasets. The experimental results validate the reliability of the proposed algorithms. The results also show that our algorithms outperform or are at least comparable to the competing online AUC maximization methods.